Open sofiasanz opened 4 years ago
I think you should do this like I do in fermi_level
i.e. pass q
as a float or a list of floats. If only one value is passed, it is the total number, otherwise it is per-spin.
In fact this is something I think would be generally favourable in the implementation. Instead of doing n_up=A, n_dn=B
one could do q=[A, B]
and fail if the list does not have length 2.
Sounds like a good idea with q=q_tot
or q=[q_up, q_dn]
as acceptable inputs.
Allow
iterative3
method for either 1) convergence towards a total number of electrons, and 2) towards a user-defined spin-configuration in open-systems.